ASCE Geotechnical Special Publication (GSP) No. 93 - National Geotechnical
Experimentation Sites (NGES), Ed. by Jean Benoit & Alan Lutenegger, ASCE,
Reston/VA, April 2000, pp. 160-185.
SITE CHARACTERIZATION OF PIEDMONT RESIDUAL SOILS
AT THE NGES, OPELIKA, ALABAMA
Paul W. Mayne1, M. ASCE, Dan Brown2, M. ASCE, James Vinson3, A.M. ASCE,
James A. Schneider4, A.M. ASCE, and Kimberly A. Finke5, A.M. ASCE
Abstract
The National Geotechnical Experimentation Site at Opelika, Alabama is underlain
by residual silts and sands derived from the weathering of schist and gneiss of the
Piedmont Geologic Province. Site characterization by in-situ tests has included an
extensive series of cone and piezocone penetrometer soundings supplemented with
standard penetration, flat plate dilatometer, pre-bored and full-displacement
pressuremeter, and borehole shear. Geophysical surveys have been completed by
crosshole, surface wave, and downhole tests. Complementary laboratory testing on
split-barrel drive samples and continuous geoprobe samples have consisted of index,
moisture, plasticity, gradation, & hydrometer analyses. Undisturbed tube samples
have been subjected to triaxial, resonant column, permeability, and oedometer
loading. Piedmont residual soils are unusual in that they exhibit behavioral facets
characteristic of both clay and sand, thus creating a dichotomy in practice in
selecting undrained versus drained soil parameters for analysis.
1. Assoc. Professor, School of Civil & Environmental Engrg, Georgia Institute of
Technology, Atlanta, GA 30332-0355 (Email: [email protected]).
2. Assoc. Professor, Dept. of Civil Engineering, Harbert Engrg. Center, Auburn
Univ., Auburn, AL 36849-5337 (Email: [email protected]).
3. Geotechnical Engineer, Soil Testing Engineers Inc., 4100 Louisiana Ave., Lake
Charles, LA. 70607.
4. Staff Engineer, GeoSyntec Consultants, 1100 Hearn Drive, Suite 200, Atlanta,
GA
30342-1523. (Email: [email protected]).
5. Geotechnical Engineer, URS Greiner Woodward Clyde, Stanford Place 3, Suite
1000, 4582 S. Ulster St., Denver, CO 80237. (Email: [email protected]).
1 - Mayne, Brown, Vinson, Schneider, & Finke
Geologic Setting
Piedmont residuum serves as the foundation bearing soils for several major cities in
the eastern United States, including Atlanta/GA, Columbia/SC, Charlotte/NC,
Raleigh/NC, and Gaithersburg/MD, as well as western suburbs of Philadelphia/PA,
Washington/DC, Wilmington/DE, Baltimore/MD, and Richmond/VA. Due to the
importance of this geology with respect to geotechnical practice in the Atlantic
region, the Opelika Test Site was established at the southern end of the Piedmont
Province, as shown by Figure 1. The exposed surface extent of the Piedmont
geologic province is 1200 kilometers long and up to 200 kilometers wide, appearing
as a lenticular formation running parallel with the Atlantic Coast. At the Fall Line,
the Piedmont actually dips easterly and underlies the extensive sediments of the
Atlantic Coastal Plain. The name Piedmont means “foot of the mountains”,
reflecting a highly eroded geomorphological state since rolling terrain is the now the
remnant of what once were mountains. The original Paleozoic rocks were primarily
of metamorphic and igneous origin, including predominances of schist, gneiss, and
granite, although localized regions contain slate, phyllite, greenstone, and diabase.
The rocks of the Piedmont have decomposed mechanically and chemically inplace
to form residuum and saprolite.
Figure 1. Surface Extent of Piedmont Geology and Opelika NGES Location.
2 - Mayne, Brown, Vinson, Schneider, & Finke
Figure 2. Weathering Profile in Residuum Derived From Piedmont
Metamorphic Rock Types (modified after Sowers & Richardson, 1983).
Residual soils transition into a saprolite and disintegrated or partially-weathered
rock with depth until refusal is encountered at the parent rock interface (Martin,
1977; Sowers & Richardson, 1983), as depicted in Figure 2. The Piedmont residual
soils are not particularly well-categorized by the Unified Soil Classification System
(USCS). The USCS was based on extensive testing and empirical understanding of
more extensive and well-studied sedimentary deposits from marine, lacustrine,
alluvial, fluvial, and glacial origins. The surface soils (< 1 m) may have severely
weathered to fine-grained silt or clay, thus in the southern Piedmont it has been
called “Georgia red clay” because of its agricultural relationships. However, the clay
fraction of Piedmont soils is often small (< 10 %) and most index tests indicate
nonplastic material, thus “red clay” is a misnomer.
Using the USCS, a vertical profile in the Piedmont appears as if alternating
strata of silty sands (SM) and sandy silts (ML) form the overburden. The strata seem
to alternate in random fashion, thus suggesting high variability over short distances.
This is illusionary and due to the fact that the mean grain size of the Piedmont
residuum is close to the opening size of the U.S. No. 200 Sieve (e.g., 0.075 mm). In
fact, the residuum acts more as a dual soil type (SM-ML), exhibiting characteristics
of both fine-grained soils (undrained) and coarse-grained soils (drained) when
subject to loading. Due to differential weathering over horizontal distances, the
residual soil overburden portion of the Piedmont can be thin (< 3 m) to over 50
meters thick.
3 - Mayne, Brown, Vinson, Schneider, & Finke
Opelika Test Site
A national geotechnical experimentation site (NGES) was established near the
small village community of Spring Villa, south of I-85 and east of Opelika, Alabama
(Vinson, 1997). The site is contained within the southern Piedmont geology and
located at the northeastern edge of the Opelika NGES property. The subsurface
materials are composed of silty to sandy residual soils that grade eventually with
depth into partially-weathered schist and gneiss. The water table lies about 2 to 3 m
deep with the upper vadose zone above the phreatic surface appearing desiccated and
crustal due to past groundwater fluctuations.
The Opelika NGES was established by the Alabama DOT as a 130-hectare tract
(320 acres) for research projects related to bridge pier foundations and pavement
subgrades. The original property consisted of forested rolling terrain. The site is
managed by Auburn University and includes a deep soils site (discussed herein) at
the south end of the property which is in use for full-scale load tests on deep
foundations under axial and lateral static and dynamic loading. Ongoing research
at the site includes construction of a racetrack for subgrade evaluations and
performance of deep foundations in weathered rocks. At the soil test portion of the
NGES, a variety of laboratory and field tests have been performed to characterize the
Piedmont residuum, as discussed subsequently.
SPT and Index Results
A summary of standard penetration test (SPT) resistances from four soil test
borings using hollow stem augers is presented in Figure 3a. Below a crustal zone
evident in the upper 3 meters, the recorded N-values increase from about 8
blows/300 mm at z = 3 m to 14 blows/300 mm at the termination depths z = 15 m.
Similar results with N-values increasing with depth were obtained at an earlier test
site in Piedmont soils in Atlanta (Harris & Mayne, 1994).
A summary of mean grain size (D50) from mechanical and hydrometer analyses
is shown in Figure 3b. In certain specimens, the gradation tests were stopped at the
percent fines content (No. 200 sieve), so that the D50 was not determined for several
of the fine-grained samples. In any event, it can be seen that the mean grain size of
the Piedmont soils is near the USCS cutoff of 0.075 mm (U.S. No. 200 sieve), thus
borderline in its classification as sandy silt to silty sand. Companion lab tests gave
a mean liquid limit of 46% and mean plasticity index of 8%, although a number of
specimens were nonplastic (Brown & Vinson, 1998). Thus, the Piedmont residual
soils are better described by dual symbols (SM-ML).
Geophysical Surveys
Crosshole and surface wave measurements were performed at the Opelika NGES to
profile the shear wave velocity with depth. Two crosshole arrays were installed at
4 - Mayne, Brown, Vinson, Schneider, & Finke
0
0
2
2
4
4
B-1
B-2
6
Mean
B3
8
10
B4
12
B6
B-3
Depth (meters)
Depth (meters)
B1
6
B-4
8
B-5
10
B-6
12
14
14
16
16
0
5
10
15
20
N-Value (blows/0.3 m)
0.01
Fine
Coarse
B-7
B-8
0.1
1
Mean Grain Size (mm)
GP
Figure 3. Summary Profiles of (a) SPT Resistances and (b) Mean Grain Sizes.
the site with each set comprised of three PVC-cased boreholes to 15 meters depth.
The boreholes in each array were about 3 meters apart. The casing was grouted after
placement using a cement-bentonite mixture and an inclinometer was used to
determine the vertical alignment of each borehole (Brown & Vinson, 1998). A
downhole hammer was used as a source in one outer borehole with receiver
geophones placed at the same level in the middle and farthest borehole to allow both
direct source-to-receiver, as well as interval receiver-receiver, determinations of
shear wave velocity (Vs), per ASTM D-4428. Results from both crosshole series
were similar (Kates, 1996) with the profile from one of the tests (CHT-2) presented
in Figure 4. Generally, in the upper 11 meters at the NGES, the Vs is rather constant
at 200± m/s.
The method of spectral analysis of surface waves (SASW) was performed using
a two-channel HP-spectrum analyzer and paired set of geophones that was moved
at the surface to obtain the dispersion curve at the site. Details of the test method,
data collection, and inversion processing are given by Rix (1998). The measured
surface waves (or Rayleigh waves) are a special type of shear wave that occurs
because of the boundary condition formed by the plane of the ground surface.
Results from the
5 - Mayne, Brown, Vinson, Schneider, & Finke
Figure 4. Results of Crosshole and Surface Wave Measurements at Opelika.
SASW are also presented in Figure 4 and show to be in general agreement with the
crosshole tests below the desiccated region. Subsequently, results from downhole
Vs measurements taken during seismic cone and seismic dilatometer tests are
presented, as well as laboratory resonant column tests on recovered samples, with all
tests indicating comparable results (Mayne, 1999).
Cone Penetration Tests
Several series of cone penetration tests (CPTs) have been conducted at the
Opelika test site since 1995. The soundings have been advanced using a variety of
penetrometers and cone rigs. Figure 5 shows one of four rigs used at the Opelika
NGES, including: (a) 20-tonne Hogentogler truck, (b) 25-tonne Fugro truck, (c) 20tonne track-mounted van den Berg rig; and (d) 6-tonne Geostar truck with twin earth
anchors. The latter is operated by the In-Situ Testing research group at Georgia
Tech. The soundings have been performed in general accordance with ASTM D
5778 guidelines using both 35.7- and 44-mm diameter penetrometers advanced at the
standard rate of 20 mm/s, although accelerated rates have also been investigated
(Finke, 1998). The various types of penetrometers (Fugro, Hogentoger, van den
Berg, and Vertek) all provided readings of cone tip stress (qc), sleeve friction (fs), and
6 - Mayne, Brown, Vinson, Schneider, & Finke
Figure 5. Cone Rig at the NGES, Opelika, AL
penetration porewater pressure (u) at regular depth intervals of 2 or 5 cm. In certain
soundings, the vertical inclination of the penetrometer (i) was also monitored during
penetration and specialized penetrometers provided other types of soil measurements
(e.g., conductivity, dielectric, downhole seismic velocities).
Piezocone Tests
For the piezocone penetration tests (PCPT or CPTu), the investigations included
penetration porewater pressure readings taken using porous filters located at either
the mid-face (u1) and/or shoulder (u2) position (Finke, et al. 1999). In general, either
a 50/50 mixture of glycerine and water, or else pure glycerine, was used to saturate
the elements and cavities. Midface filters were made of porous plastic (Hogentogler)
and ceramic (Fugro). Shoulder elements were made of porous plastic for all types.
7 - Mayne, Brown, Vinson, Schneider, & Finke
0
0
0
-200
0
2
2
2
2
4
4
4
4
6
6
6
6
8
8
0
Depth (m)
u2 (kPa)
u1 (kPa)
fS (kPa)
qT (MPa)
8
2
4
6
8
0
8
200
F-1
(u1)
F-2
(u2)
0
400
500
1000
0
200
400
uo
10
10
10
10
Figure 6. Piezocone Test Results from Two Sister Soundings Showing Both
Midface (u1) and Shoulder (u2) Porous Filter Responses.
Results from a paired set of 15-cm2 soundings side-by-side with the two types
of pore pressure elements are shown in Figure 6. The tip stresses have been
corrected as per recommended practice (Campanella & Robertson, 1988; Lunne, et
al. 1997), although the correction from qc to qt is not significant in these residual soils
(Finke, et al. 1999). The cone tip stresses measure about 2.5 < qt < 3.5 MPa in the
upper 10 meters. Corresponding sleeve frictions are between 150 < fs < 250 kPa.
For penetration porewater pressures, dramatically different responses are evident
with the midface readings high and positive (400 < u1 < 800 kPa) and the shoulder
element readings negative and near cavitation (u2 . -90 kPa). This phenomenon of
positive u1 with negative u2 response has been previously observed as characteristic
of fissured overconsolidated clays (Mayne, et al. 1990; Lunne, et al. 1997).
The Piedmont residuum has relict features and qualities of the parent rock,
including remnant bonding of the intact rock itself, as well as the discontinuities,
jointing, cracks, and fissures of the rock mass (Sowers, 1994). It is postulated that,
8 - Mayne, Brown, Vinson, Schneider, & Finke
since the midface u1 filter lies within a zone of high compression located directly
beneath the cone tip, this response is positive because it is dominated by the
destructuration of the residual intact bonding of the original relict rock continuum.
In contrast, the shoulder u2 readings are negative because they reflect shear-induced
pore pressures and remnant discontinuities within the matrix (St. John, et al. 1969).
Thus, the u1 readings reflect the destructuration of intact bonds and the u2 readings
are indicative of the fissures, fractures, and jointing of the original rock mass (Finke
& Mayne, 1999).
In the vadose zone above the groundwater table, the penetration porewater
pressures at midface have been observed to be either positive or “zero-ish”, while at
the shoulder, the values can be positive, zero, or negative. It is believed that these
reflect the transient capillary conditions due to the current degree of partial or full
saturation in the residual fine-grained soils which depends on the humidity,
infiltration, and prior rainfall activities around the actual time of testing.
Porewater Pressure Dissipations
During selected piezocone soundings, a series of porewater pressure dissipation tests
were performed to quantify the horizontal coefficient of consolidation and
permeability characteristics. Figure 7 shows six sets of dissipation results obtained
Porewater Pressure (kPa)
600
Depth
500
Midface u1
400
5.59 m
300
u2
u1
200
6.57 m
7.47 m
100
uo
0
8.38 m
-100
10.45 m
Shoulder u2
-200
1
10
100
Time (seconds)
1000
13.53 m
Value at Penetration
Figure 7. Representative Porewater Dissipations to Hydrostatic Conditions.
9 - Mayne, Brown, Vinson, Schneider, & Finke
using a special 10-cm2 dual-element penetrometer capable of monitoring u1 and u2
simultaneously. At depths ranging from 5.6 m to 13.5 m, separate paired records of
pore pressure decay versus logarithm of time are shown. For reference, the
associated values of u1 and u2 at the time of penetration are shown at t = 1 second.
All dissipations reached hydrostatic values (u0) and corresponded to the ambient
unconfined water table at 3 m. Despite the vast differences in u1 and u2 readings
taken during penetration, full dissipation to the same equilibrium value was achieved
at each depth in only 0.5 to 2 minutes for both midface and shoulder elements.
Figure 8 shows the initial values of penetration porewater pressures and final values
at equilibrium, corresponding to hydrostatic conditions. The quick time for 100%
decay indicates fairly high values of coefficient of consolidation (ch . 0.63 cm2/s)
and hydraulic conductivity (k . 1 A 10-4 cm/s) for the silty to sandy Piedmont soils
at Opelika (Finke, 1998).
Porewater Pressure (kPa)
-100
0
2
0
100
200
300
400
u dissipation
1
u dissipation
2
4
Depth (meters)
6
8
10
12
u during
1
penetration
14
u during
2
penetration
16
Figure 8. Summary of Initial and Final Porewater Pressures from Piezocones.
10 - Mayne, Brown, Vinson, Schneider, & Finke
Seismic Cone Tests
Several CPTs have been performed at the Opelika NGES using a seismic
penetrometer that includes a single horizontal geophone (velocity transducer) within
the upper section of the probe. Using a horizontally-struck plank that is oriented
parallel with the geophone axis, a vertically-propagating horizontally-polarized shear
wave can be monitored at one-meter depth intervals, usually corresponding to the
successive addition of cone rods. The source is usually located one to two meters
from the axis of the vertical sounding. This downhole geophysical technique
provides a pseudo-interval determination of Vs with depth (Campanella, 1994).
Striking the plank from the left and then from the right produces two separate waves
that are mirrored images. These can be superimposed to accentuate the arrival time
of the shear wave on the oscilloscope screen and provide a better delineation during
interpretation.
Figure 9 presents the results of a seismic piezocone penetration test (SCPTu)
which conveniently provides four independent readings with depth from a single
sounding (qt, fs, ub, and Vs). The downhole Vs were relatively constant at 200 ± 50
m/s in the upper 12 meters and compare well with the values obtained from the
aforementioned CHT and SASW surveys(Schneider et al. 1999).
qt (MPa)
Depth (meters)
0
2
4
fs (kPa)
6
8
0
100
200
Vs (m/s)
u2 (kPa)
300
0
100
0
0
0
-100
0
200
2
2
2
2
4
4
4
4
6
6
6
6
8
8
8
8
10
10
10
12
12
12
200
400
0
uo
10
12
SCPTu
SDMT
Crosshole
SASW
Figure 9. Representative Seismic Piezocone Sounding at Opelika, Alabama.
11 - Mayne, Brown, Vinson, Schneider, & Finke
Flat Plate Dilatometer Tests
The flat plate dilatometer test (DMT) provides measurements of contact pressure
(po) and expansion pressure (p1) at regular intervals of either 20-cm or 30-cm with
depth. These readings are used to evaluate the stratigraphy and strength properties
of the soils. At Opelika, both the regular blade unit and a special modified system
outfitted with a geophone in the rod coupler have been used (Martin & Mayne,
1998). Results from one regular DMT and two seismic dilatometer tests (SDMT) are
given in Figure 10. The apparent crustal and desiccated zone is evident in the upper
three meters in all three measurements (po, p1, Vs). Beneath this, the readings are
seen to increase slightly with depth. The shear wave data can be used to provide an
initial stiffness, represented by the small-strain shear modulus, G0 = D(Vs)2, where
D = total mass density of the soil. The DMT pressures can be used to evaluate the
soil strength. Thus, it is possible to construct an entire stress-strain-strength curve
at each depth from the results of the SDMT, as discussed elsewhere (Mayne,
Schneider, & Martin, 1999).
Depth (meters)
Vs (m/s)
p1 (bars)
po (bars)
0
0
0
2
2
2
4
4
4
6
6
8
8
8
10
10
10
SDMT-AU1
SDMT-AU2
CHT-SR2
6
DMT-99
12
12
0
5
Contact
10 0
5
10
Expansion
12
15 0
200
400
Shear Wave
Figure 10. Results from Seismic Flat Plate Dilatometer Tests at Opelika NGES.
Dissipation tests can also be performed using the flat plate dilatometer
(Lutenegger, 1988). A special series of DMT dissipations tests have been carried
out at Opelika using the A-readings (unpublished). The latter indicate the Areadings decay to about 90% of their penetration value after 2 minutes. A separate
series of DMTs has been performed whereby the A and B readings were recorded 2minutes after the blade was installed.
12 - Mayne, Brown, Vinson, Schneider, & Finke
Pressuremeter Tests
Menard pre-bored type pressuremeter tests (PMT) were conducted at the NGES
using a monocell “Texam”-type probe. The hole was pre-formed by first pushing a
71-mm diameter thin-walled Shelby tube sample at the bottom of the borehole, prior
to placement of the pressuremeter. A total of 16 PMTs were completed (Brown &
Vinson, 1998). Sample results for the expansion pressure versus measured volume
curves are given in Figure 11 from two boreholes at two test depths. Details on the
assessment of total horizontal stress (Po), elastic modulus (Epmt), strength, and limit
pressure (PL) are discussed in Vinson (1997).
Expansion Pressure, P (kPa)
1000
800
600
400
200
0
0
200
400
600
Corrected Volume, V (cc)
B2 at 8 m
B5 at 8 m
B2 at 10 m
B5 at 10 m
Figure 11. Representative Results from Pre-Bored Type Pressuremeter Tests.
Full-displacement type pressuremeter tests (FDPMT) were also conducted at
Opelika using a 35.7-mm diameter Pencel probe having a 60/ apexed-tip at its front
end. This probe is pushed directly into place, thus an assessment of in-situ lateral
stress state is not obtainable. However, the test is considerably quicker than regular
PMTs that took two days to complete, compared with a total of 29 FDPMTs which
were completed in only 6 hours work. Selected results at three test depths are
shown in Figure 12, including an unload-reload cycle to better define some pseudoelastic stiffness of the Piedmont residual soils.
13 - Mayne, Brown, Vinson, Schneider, & Finke
Expansion Pressure, P (kPa)
1000
800
600
400
200
0
0
20
40
60
Measured Volume, V (cc)
C42P at 6 m
C42P at 8 m
C42P at 10 m
Figure 12. Representative Data from Full-Displacement Pressuremeter Tests.
Figure 13. Conducting CPTs using anchored rig at Opelika, Alabama
Borehole Shear Tests
Iowa borehole shear (IBS) tests were performed downhole in selected soil test
borings using a stage-loading procedure. Details are given elsewhere (Vinson &
Brown, 1997). These IBS results showed considerable variations in the derived
Mohr-Coulomb strength parameters.
14 - Mayne, Brown, Vinson, Schneider, & Finke
Laboratory Testing Program
In addition to drive samples, thin-walled Shelby tube samples were readily
obtainable in the Piedmont residual silts and sands. At the Opelika NGES, several
sets of tube samples have been retrieved and subjected to a varied assortment of tests,
including: unit weight determinations, one-dimensional consolidation, permeability,
drained & undrained triaxial shear, and resonant column.
Oedometer Series
The e-log Fvr curves derived from one-dimensional consolidation tests on
Piedmont soils do not show a clear yield point (see Figure 14), thus it is difficult to
assign any reputable value of preconsolidation stress, or corresponding
overconsolidation ratio (OCR) to these materials (Wesley, 1994). Prior efforts at
defining the preconsolidation stress in these materials have been attempted by Pavich
& Obermeier (1985) who indicated Piedmont soils to be normally-consolidated.
Mayne & Harris (1993) suggested the apparent overconsolidation ratios (AOCR) of
Piedmont silty sands were between 1.5 and 2.5 within the upper 15 m in downtown
Atlanta/GA, while Wang & Borden (1996) showed 3 < AOCR < 4 for a site east of
Raleigh/NC. At the Opelika site, Hoyos & Macari (1999) evaluated AOCRs
decreasing from 4 to 1.1 in the upper 9 meters.
2.0
1.8
1.6
Void Ratio, e
1.4
1.2
1.0
0.8
B9 - 3 meters
0.6
B9 - 4 meters
0.4
B7 - 5 meters
B8 - 7 meters
0.2
B7 - 9 meters
0.0
1
10
100
1000
10000
Effective Consolidation Stress (kPa)
Figure 14. Oedometer Test Data From Piedmont Residuum.
15 - Mayne, Brown, Vinson, Schneider, & Finke
Since these geomaterials have formed inplace from the debonding of the
underlying parent rock, the mechanism is one that is opposite of the traditional
maximum applied prestressing associated with increased overburden and subsequent
erosion or glaciation (e.g. loading-unloading). The Piedmont soils are very old and
it is likely that any any preconsolidation effects can likely be attributed to more
recent desiccation and groundwater fluctuations that have occurred since the primary
debonding events. Expected OCRs due to these latter mechanisms are in the range
of 1.5 to 4.
Triaxial Shear Tests
A series of isotropically-consolidated undrained and drained triaxial shear tests
were conducted to define the strength and stiffness parameters of the Piedmont
residuum at the NGES (Brown & Vinson, 1998).
Figure 15 shows the
corresponding stress paths in terms of the parameters q = ½(F1 + F3) and pr = ½(F1rF3r), as well as the interpreted Mohr-Coulomb parameters (cr = 17 kPa; Nr = 31/).
The majority of tests were CIUC type and overall showed little excess porewater
pressure development. No clear trend of contractive nor dilative behavior was
observed for the Opelika soils, as noted previously for Piedmont soils in Atlanta
(Mayne & Harris, 1993). If the effective cohesion intercept is taken to be negligible
(cr = 0), then Figure 16 shows the alternative interpretation for the stress path points
at failure in MIT q-p’ space (Nr = 35.3/).
Figure 15. Laboratory Equipment for Consolidated Triaxial Tests
Corresponding total stress parameters can also be defined for the Piedmont soils.
Results from laboratory UU and CIUC tests determined an average su = 92 kPa
within the upper 15 m. Comparisons with values from SPT, DMT, CPT, and PMT
16 - Mayne, Brown, Vinson, Schneider, & Finke
500
Mohr-Coulomb Parameters:
c' = 0; Phi' = 35.3°
MIT-q (kPa)
400
300
y = 0.5782x
R2 = 0.9258
200
100
0
0
100
200
300
400
500
MIT-p' (kPa)
Figure 16. Frictional Envelope of Piedmont Residual Soils at Opelika.
are discussed elsewhere (Vinson & Brown, 1997). Interestingly, a fairly high value
for the cone bearing factor of NKT = 36 was required to match the CPT results with
the laboratory data, whereas commonly adopted values for intact clays are often
taken in the range: 8 < NKT < 16 for compression type loading modes (Lunne, et al.
1997).
Permeability
A very limited number of falling head permeability tests have been conducted on
tube samples (Finke, 1998). These tests indicated representative values of
coefficient of hydraulic conductivity (k . 1 A 10-4 cm/s) for the sandy silts to fine
sandy silts of the Piedmont geology at Opelika.
Resonant Column
A total of 12 resonant column specimens (142-mm height; 72-mm diameter solid
cylinders) were tested under isotropic consolidation using a Stokoe device. The
specimens were recovered and trimmed from 10 Shelby tubes retrieved at Opelika
NGES. The specimens corresponded to residual soil formations (SM and ML) at
depths ranging from 3 m to 10 m. Details are presented in Schneider et al. (1999)
and Hoyos & Macari (1999).
17 - Mayne, Brown, Vinson, Schneider, & Finke
Two samples per depth were tested in the resonant column device for validation
of test results. Each specimen was tested at five different levels of confinement
corresponding to 0.5, 1, 2, 4, and 8 times the effective overburden stress. This range
of confining pressures was selected to reflect states of stress below, near, and above
the in-situ stress state. Figure 17 presents the summary of small-strain shear moduli
(Gmax or G0) versus effective confining stress (Fcr). Both axes are made
dimensionless by dividing by a reference pressure equal to one atmosphere (pa = 1
bar . 100 kPa . 1 tsf).
Figure 17. Laboratory Resonant Column Test Setup for Obtaining SmallStrain Shear Modulus (Gmax) of Piedmont Residual Soils from Opelika AL.
Discussion
A comparison of the strength and stiffness parameters from the various in-situ
and lab tests has been made and presented by Brown & Vinson (1998). Since shear
modulus (G) is highly dependent on level of shear strain, Figure 18 indicates a very
wide variation in measured values, from as low as 2 MPa for the initial loading curve
of a Menard PMT to as high as 120 MPa for the small-strain value obtained in field
geophysical surveys. Intermediate values of G are obtained with the flat DMT, lab
triaxial, and unload-reload cycles applied during the full-displacement type PMT.
18 - Mayne, Brown, Vinson, Schneider, & Finke
0
Lab Triaxial Tests
2
Flat Dilatometer
4
Menard PMT
Depth (m)
6
Initial-FDPMT
8
Reload-FDPMT
10
SCPT
12
SDMT
CHT
14
16
1
10
100
1000
Shear Modulus (MPa)
Figure 18. Soil Stiffnesses from Various Tests at Opelika in Terms of Shear
Modulus (G) at Different Strain Levels. Note: Modulus E = 2G(1+<).
An interesting dichotomy occurs in the evaluation of total stress versus effective
stress parameters at Opelika. Although excess porewater pressures occur during
penetration, the interpretation of SPT, CPT, and PMT by conventional analyses seem
to provide reasonable values of effective stress friction angle (Nr), thus supporting
the favor of interpreted drained behavior in these residual soils. Figure 19 illustrates
the utilization of the SPT method (Schmertmann, 1975), CPT data (Kulhawy &
Mayne, 1990), and recent DMT approach by Marchetti (1997). The SPT N-values
were taken from borings numbered B1, B3, and B4 (see prior Figure 3). For the
comparison presented herein, the cone tip resistance (qt) from CPT sounding number
CPT42 (previously reported by Mayne, 1999) and the value of horizontal stress
index (KD) from DMT-99 (Fig. 10) were used. As seen in Figure 19, the interpreted
Nr from these three in-situ penetration tests give comparable values to the Nr = 35/
(assuming c’ = 0) obtained from the triaxial series (prior Fig. 16). The CPT
approach by Robertson & Campanella (1983) gave Nr values about 1/ higher at
shallow depths and -1/ lower at deeper depths. In addition, the interpreted values of
effective stress friction angles for the full-displacement and prebored PMT data are
30/ < Nr < 33/ (Vinson & Brown, 1997).
19 - Mayne, Brown, Vinson, Schneider, & Finke
Effective Friction Angle (deg)
25
30
35
40
45
0
2
4
CPT
SPT
Depth (meters)
DMT
6
Triaxial
8
10
12
14
16
Figure 19. Interpreted Effective Stress Strength Parameters at Opelika NGES.
In contrast, a total stress interpretation of in-situ data suggests that the Piedmont
residuum acts as a stiff fissured clay. Evidence for this is three-fold: (1) penetration
porewater pressure response; (2) backcalculated cone bearing factor; (3) shear wave
data; as explained subsequently.
Reviews of piezocone data worldwide have showed that the unusual
combination of penetration porewater pressures (positive u1 yet negative u2) occur
in stiff fissured geomaterials (Mayne, Kulhawy, & Kay, 1990; Lunne, Robertson,
& Powell, 1997). Examples include the fissured clays in London (Brent Cross),
Gault clay (Madingley), and Beaumont formation (Texas). The positive u1 readings
are likely indicative of destructuration of the relict bonding of the parent intact rock,
yet the negative u2 readings reflect the remnant cracks, fissures, and discontinuities
of the former rock mass (Finke, et al. 1999). Both facets are still represented within
the residual soil matrix. Mapped evidence of pre-existing cracks, fissures, and
slickenslides in the Piedmont has been documented previously (St. John, et al. 1969;
Sowers & Richardson, 1983).
Because of the relict presence of fissures & cracks, the operational undrained
strength is less than that of a comparable intact “clay”. Thus, prior work has shown
that backcalculated NkT factors for fissured materials are on the order of 25 to 30+
(Marsland & Powell, 1988; Powell & Quarterman, 1988), thus in agreement with the
20 - Mayne, Brown, Vinson, Schneider, & Finke
values obtained by Brown & Vinson (1998). It is important to cite the specific mode
of undrained shear under consideration (i.e., triaxial, simple shear, plane strain) and
direction of loading (i.e., compression vs. extension), as the undrained shear strength
(su) depends significantly on these facets, as well as the effects of strain rate,
anisotropic stress state, and other factors (Kulhawy & Mayne, 1990).
Figure 20 shows the undrained shear strengths (suTC) for triaxial compression
loading from reference laboratory CIUC tests compared with in-situ interpretations
based on the CPT using a bearing factor Nkt = 35 (Lunne, et al. 1997), SPT (Stroud,
1988), and DMT with Nc = 7 (Roque, et al. 1988). The overall average suTC from
UU and CIUC tests is 92 kPa at the Opelika site. The triaxial data shown herein are
only CIUC results performed at the current effective vertical overburden stresses.
Surprisingly, the same SPT, CPT, and DMT data were used to generate both the total
stress parameters (Fig. 20) and effective stress parameters (Fig. 19), thus illustrating
the hermaphroditic strength characteristics of the Piedmont residuum.
Undrained Shear Strength for TC (kPa)
0
50
100
150
200
250
300
0
2
CPT
Depth (meters)
4
SPT
6
DMT
8
Triaxial
10
12
14
TC = Triaxial
Compression M ode
16
Figure 20. Interpreted Undrained Shear Strengths at Opelika NGES.
21 - Mayne, Brown, Vinson, Schneider, & Finke
Interrelationships between small-strain G0 and penetration resistance qT have
been formulated for different soil types, including clean quartz sands (Baldi, et al.
1989), clays (Mayne & Rix, 1993), as well as a full range of soil types (Lunne, et al.
(1997). Here too, a dichotomy occurs in that the use of the seismic piezocone data
(taken from Fig. 9) presented in terms of normalized cone tip resistance [Q = (qtFvo)/Fvor] versus the ratio G0/qt plots within the region of “sand mixtures” grading
to “silt mixtures”, as shown by Figure 21. This gives an apparent reasonable
classification according to soil type by gradation and is consistent with the
aforementioned dual ML/SM nomenclature. However, when the same data are put
in terms of the clay database (Figure 22) and presented directly as G0 vs. qt, the
results suggest that the Piedmont soils are behaving analogous to stiff fissured clays.
Normalized Q = (qt-σvo)/σvo'
1000
Gravelly
Sands
Piedmont
Residuum
100
Sand
Sand
Mixtures
10
Silt
Mixtures
1
1
Clays
Young
Uncemented
Soils
10
100
Ratio Go/qt
Figure 21. Normalized Cone Tip Stress Q vs. Ratio G0/qt from SCPTu
Sounding. (Soil Behavioral Classification System given in Lunne, et al. 1997).
22 - Mayne, Brown, Vinson, Schneider, & Finke
Figure 22.
Small-Strain Stiffness (G0) versus Cone Tip Stress (qt) in Intact to
Fissured Clay Geomaterials, including Piedmont SCPTu Data.
Conclusions
A geotechnical characterization program has been directed at the improved
understanding of Piedmont residual soils which are comprised of fine sandy silts and
silty fine sands derived from weathered gneiss and schist. At the Opelika NGES in
southeastern Alabama, several series of in-situ field penetration tests (SPT, CPT,
PCPT, SCPT, DMT, SDMT, FDPMT), borehole tests (PMT, IBS), and geophysical
techniques (CHT, DHT, SASW) have been performed and complemented by series
of laboratory tests (index, triaxial, consolidation, resonant column). Piedmont soils
are unusual in that they exhibit behavioral features which are characteristic of both
sand and clay.
Acknowledgments
The authors appreciate the funding provided by the National Science
Foundation, Mid-America Earthquake Center, and the Alabama DOT, as well as
support and assistance provided by Fugro Geosciences, Morris-Shea Bridge Co.,
Hogentogler & Company, AMS Inc., Vertek, Univ. of Florida, and Williams Earth
Sciences.
23 - Mayne, Brown, Vinson, Schneider, & Finke
References
Baldi, G., Bellotti, R., Ghionna, V., Jamiolkowski, M., and LoPresti, D.C.F. (1989).
Modulus of sands from CPT and DMT. Proceedings, 12th International
Conference on Soil Mechanics & Foundation Engrg., Vol. 1, Rio, 165-170.
Borden, R.H., Shao, L, and Gupta, A. (1996). Dynamic properties of Piedmont
residual soils. Journal of Geotechnical & Geoenvironmental Engineering, Vol.
122 (10), 813-821.
Brown, D.A. and Vinson, J. (1998) Comparison of strength and stiffness parameters
for a Piedmont residual soil, Geotechnical Site Characterization, Vol. 2, (Proc.
ISC’98, Atlanta), Balkema, Rotterdam, 1229-1234.
Campanella, R.G. and Robertson, P.K. (1988). Current status of the piezocone test,
Penetration Testing 1988, Vol. 1, Balkema Publishers, Rotterdam, 93-116.
Campanella, R.G. (1994). Field methods for dynamic geotechnical testing. Dynamic
Geotechnical Testing II, (STP 1213), ASTM, West Conshohocken, 3-23.
Finke, K.A. (1998). Piezocone penetration testing in Piedmont residual soils, MS
Thesis, School of Civil and Environmental Engineering, Georgia Institute of
Technology, Atlanta, 265 p.
Finke, K.A. and Mayne, P.W. (1999). Piezocone tests in U.S. Atlantic Piedmont
residual soils. Proceedings, XI Pan American Conference on Soil Mechanics &
Geotechnical Engineering, Vol. 1, Foz do Iguassu, Brazil, 329-334.
Finke, K.A., Mayne, P.W., and Klopp, R.A. (1999). Characteristic piezocone
response in Piedmont residual soils. Behavioral Characteristics of Residual
Soils, GSP No. 92, ASCE, Reston, VA, 1-11.
Harris, D.E., and Mayne, P.W. (1994). Axial compression behavior of two drilled
shafts in Piedmont residual soils, Proceedings, International Conference on
Design and Construction of Deep Foundations, Vol. 2 (Orlando), Federal
Highway Administration, Washington, D.C., 352-367.
Hoyos, L., Jr. and Macari, E.J. (1999). Influence of in-situ factors on dynamic
response of Piedmont residual soils. Journal of Geotechnical and Geoenvironmental Engineering, Vol. 125 (4), 271-279.
Kulhawy, F.H. and Mayne, P.W. (1990). Manual on Estimating Soil Properties for
Foundation Design, Report EL-6800, Electric Power Research Institute, Palo
Alto, 306 p.
Lunne, T., Powell, J.J.M., and Robertson, P.K. (1996). Use of piezocone tests in
non-textbook materials, Advances in Site Investigation Practice, Thomas Telford,
London, 438-451.
Lunne, T., Robertson, P.K., and Powell, J.J.M. (1997). Cone Penetration Testing in
Geotechnical Practice, Blackie Academic & Professional, Chapman and Hall,
London, 312 p.
Lutenegger, A.J. (1988). Current status of the Marchetti dilatometer test.
Penetration Testing 1988, Vol. 1, A.A. Balkema, Rotterdam, 137-155.
Marchetti, S. (1997). The flat dilatometer: design applications. Proceedings, Third
Geotechnical Engineering Conference, Cairo University, Egypt, 1-25.
Marsland, A. and Powell, J.J.M. (1988). Investigation of cone penetration test in
British clay. Penetration Testing in the U.K., Thomas Telford, London, 209-214.
24 - Mayne, Brown, Vinson, Schneider, & Finke
Martin, G.K., and Mayne, P.W. (1998). Seismic flat dilatometer tests in Piedmont
residual soils, Geotechnical Site Characterization, Vol. 2, Balkema Rotterdam,
837-843.
Martin, R.E. (1977). Estimating foundation settlements in residual soils, Journal of
the Geotechnical Engineering Division, ASCE, Vol. 103 (GT3), 197-212.
Mayne, P.W. (1999). Site characterization aspects of Piedmont residual soils in
eastern U.S. Proceedings, 14th Intl. Conf. on Soil Mechanics & Foundation
Engineering, Vol. 4, Balkema/Rotterdam, 2191-2195.
Mayne, P.W. and Dumas, C. (1997). Enhanced in-situ geotechnical testing for
bridge foundation analysis, Transportation Research Record, No. 1569, National
Academy Press, Washington, D.C., 26-35.
Mayne, P.W. and Harris, D.E. (1993). Axial load-displacement behavior of drilled
shaft foundations in Piedmont residuum. Research Report No. E-20-X19 to
FHWA by Georgia Tech Research Corp, Atlanta, 162 p.
Mayne, P.W., Kulhawy, F.H., and Kay, J.N. (1990). Observations on the
development of porewater pressures during piezocone penetration in clays,
Canadian Geotechnical Journal, Vol. 27, No. 4, 418-428.
Mayne, P.W., Mitchell, J.K., Auxt, J.A., and Yilmaz, R. (1995). U.S. national report
on CPT. Proceedings, Symposium on Cone Penetration Testing, Vol. 1, Swedish
Geotechnical Society, Linkoping, 263-276.
Mayne, P.W. and Rix, G.J. (1993). Gmax-qc relationships for clays. ASTM
Geotechnical Testing Journal, Vol. 16 (1), 54-60.
Mayne, P.W. , Schneider, J.A., and Martin, G.K. (1999). Small- and large-strain soil
properties from seismic flat dilatometer tests. Pre-Failure Deformation
Characteristics of Geomaterials, Vol. 1, (Proceedings, Torino’99), Balkema,
Rotterdam, 419-426.
Pavich, M.J. and Obermeier, S.F. (1985). Saprolite formation beneath coastal plain
sediments near Washington, D.C., Geological Society of America Bulletin, Vol.
96, 886-900.
Powell, J.J.M and Quarterman, R.S.T. (1988). The interpretation of cone penetration
tests in clays, with particular reference to rate effects. Penetration Testing 1988,
Vol. 2, Balkema, Rotterdam, 903-909.
Robertson, P.K and Campanella, R.G. (1983). Interpretation of cone penetration
tests in sand, Canadian Geotechnical Journal, Vol. 20 (4), 718-733.
Roque, R., Janbu, N., and Senneset, K. (1988). Basic interpretation procedures of
flat dilatometer tests. Penetration Testing 1988, Vol. 1, (Proceedings, ISOPT-1,
Orlando), Balkema, Rotterdam, 577-587.
Schmertmann, J.H. (1975). Measurement of in-situ shear strength. Proceedings, InSitu Measurement of Soil Properties, Vol. 2, (Raleigh Conference), ASCE, New
York, 57-138.
Schneider, J.A., Hoyos, L., Mayne, P.W., Macari, E.J., and Rix, G.J. (1999). Field
and lab measurements of dynamic shear modulus of Piedmont residual soils.
Behavioral Characteristics of Residual Soils (GSP 92), ASCE, Reston, 12-25.
Sowers, G.F. (1994). Residual soil settlement related to the weathering profile.
Vertical and Horizontal Deformations of Foundations and Embankments, GSP
No. 40, Vol. 2, ASCE, New York, 1689-1702.
25 - Mayne, Brown, Vinson, Schneider, & Finke
Sowers, G.F. and Richardson, T.L. (1985). Residual soils of the Piedmont and Blue
Ridge, Transportation Research Record No. 919, National Academy Press,
Washington, D.C., 10-16.
St. John, B.J., Sowers, G.F. and Weaver, C.E. (1969). Proceedings, 7th International
Conference on Soil Mechanics and Foundation Engineering, Vol. 2, Mexico,
City, 591-597.
Stroud, M.A. (1974). The standard penetration test in insensitive clays and soft
rocks. Proceedings, European Conference on Penetration Testing, Vol. 2 (2),
Stockholm, 367-375.
Townsend, F.C. (1985). Geotechnical characteristics of residual soils. Journal of
Geotechnical Engineering, Vol. 111 (1), 77-94.
Vinson, J.L and Brown, D.A. (1997). Site characterization of the Spring Villa
geotechnical test site and a comparison of strength and stiffness parameters for
a Piedmont residual soil, Report No. IR-97-04, Highway Research Center,
Harbert Engineering Center, Auburn University, AL, 385 p.
Wang, C.E. and Borden, R.H. (1996). Deformation characteristics of Piedmont
residual soils. Journal of Geotechnical Engineering, Vol. 122 (10), 822-830.
Wesley, L.D. (1994). The use of consolidometer tests to estimate settlement in
residual soil. Proceedings, XIII Intl. Conf. on Soil Mechanics & Foundation
Engrg., Vol. 2, New Delhi, 929-934.
26 - Mayne, Brown, Vinson, Schneider, & Finke